Tracking 11,709 AI launches on Hacker News (spark.site)

A recent analysis of 11,709 AI product launches on Hacker News reveals a striking trend: while the number of launches has surged (with nearly 60 daily posts by early 2026 compared to about two per day in 2022), the longevity of these products has significantly declined. Approximately 50% of the launches are now considered "dead," with newer products dying at a rate 4.2 times higher than those from 2022. The average lifespan of AI products has been shrinking, with 2022 launches lasting about 54 months compared to just 31 months for those from 2024. This rapid pace of launches—exemplified by 3,044 products introduced in February 2026 alone—coincides with an equally rapid decline in product viability, suggesting a potential saturation in the market. Interestingly, more successful launches (those receiving over 100 points on Hacker News) still experience a loss of over 25%, indicating that even public attention at launch does not guarantee long-term success. The findings highlight a critical challenge for the AI/ML community: as the barrier to entry for launching new products decreases, sustaining them becomes increasingly difficult, raising questions about market dynamics and the sustainability of innovation within this fast-evolving landscape.